End of Mindstorms?

But that is part of what Mindstorms is about. You don't know anything at first. But you start by getting a concept of what small robots are, what they can and can not do and you start learning logic in programming. After that, its pretty open with what you can do with it. Start using BrickOS and making custom sesnors.

I think many robot people take for granted what non robot people know about robots. I got friends that come over and see my robots and don't have a vauge concept of what they really are. All they know is what they see in movies. I think the mindstorms is a great opener into learning about robots.

-C

"Si Ballenger" wrote in message news: snipped-for-privacy@news.comporium.net...

I taught classes to middle school children using Mindstorms and created a whole series of challenges for them. They began to understand a lot of things, not the least of which was how delicate a Legos robot is. They began to learn about event driven programs and how to make their simple robots perform complex tasks. The key here is that they could get their hands on a simple, relatively cheap kit that allowed them to make many different machines, and that the form ad function of those machines went hand in hand. Also, they wanted more sensors, hardier components, and more instruction space in no time. Some of the things the had to learn were: 1) robots typically have no clue where they are located- getting a manipulator to a specific place or making the robot go somewhere meant having to really think about your goals and the hardware limitations 2) robots are generally not very strong- they had to design their machines for the task at hand, trading off speed and power in most circumstances 3) robots are pretty dumb and rely on what you can foresee and put into code. They had to try to plan their robots' actions and make the code anticipate whatever they would encounter. 4) the robot does not do the thinking for you- you have to be a really practical thinker to make a machine that will do just exactly what you want. In short, they learned not only about robots, but also about planning and logistics, as well as how they themselves think about things. So think of Mindstorms as a sort of catalyst- sure, it has the "toy-like" aspect, but play is how we learn, and by directing the children to certain goals, they had to constantly re-evaluate what exactly they were trying to do. A short list of some of the goals that I set out: First, we had a large board with magnet switches beneath it (reed switches). I made some Legos blocks with magnets imbedded inside epoxy that would snap under their robots. They had to make the robot navigate the board, given only distances and angles, and make each switch close, thus lighting up an indicator. They had to figure out how to make a blind, deaf, senseless box do this job reliably. Second, we had a power tractor contest where their robots had to move large blocks of aluminum. We saw that speed and power were often not compatible with each other. Each robot was of a different design, and each performed the task better or poorer depending on the tracks or wheels, gear ratios, etc. Third, I made a maze on a tabletop with black tape and the robots had to negotiate it in the shortest time. Here, there were optical sensors and fairly complex programs to make the robots stay within the boundaries. As a side note, robots were expected to not fall off the table. Fourth, I made them sweep a rectangular area clean of Styrofoam bits, and the thinking involved was very clear at the outset- every one of them made a wide sweeper for the first try, something that would get it all like a squeegee. But then I restricted the width of the sweepers, meaning that they had to make multiple passes. Now they were forced to consider how to tell their machines when they were to turn and try again, and when they were done. Fifth, I had them send IR messages from bot to bot, but each had to confirm the message it received. now they were learning a little about packet communications, if you will. Each message had to be sent in a definite sequence and had to be acknowledged properly. Sixth, we had a "lunar mission" where their robots had to count marks on the "moon", stop at a boundary, locate the "earth base" and send back the results. Keep in mind that they had to take turns being the lunar bot or the earth base, so they had to write both programs and make everything work in a room with extraneous lighting and other problems. This is a small sampler of what I did with them- I don't want to spill the whole course, as I plan to do it all again soon. We spent a week learning about robotics using Mindstorms and I believe that it was very successful. The positive feedback was enormous and the children (many of them) are now pursuing electronics, robotics, and even physics. I feel that they must have had some sort of interest to start with of course, but this helped to further it and show them that even with a toy, you can learn a lot.

Cheers!

Chip Shults My robotics, space and CGI web page -

Well, if Lego (or somebody else) sells motors and sensors, I'm sure that a better RCX could be built for around $100 to $200.

For example, for the high end, I could design a board around a Systronix JStamp

with 8 ADC and

8 motor PWM as a 1-shot for around $200, *if* I could get a plastic enclosure with the electrical connections.

I could build a board around a PIC 18F452 for around $100.

The more difficult part would be to write a simple robotics language that would work for beginners.

-- D. Jay Newman

Trying to beat the MIT system that Legos used with the Hitachi chip is going to be a real feat of programming. I think MIT spent several years coming up with it, then Legos put in their stuff too. The MIT system has that really amazing multitasking RTOS built in, and it's pretty sophisticated atually. They have that beginning user friendly front end system with the Legos block metaphore for programming. Then the more advanced people could use Mindscripts or NCP to program it as they advance. What it needs is to break out 16 I/O ports so people can plug in more than Six things to it. As it is from all the info going about, everyone was trying to double or triple up sensors on these few outputs that they had. You only had three motors and three sensors for I/O ports. Four motors and 12 I/O ports would make it a pretty neat little gadjet to go do stuff with. Of course with no marketing, and very few stores selling it, it wasn't off to a life at all. They really should have had a super RCX out a couple of years or so ago.

It sorta reminds me of other companies biting the big one with bad everything in their corporate offices. I remember Irving Gould of Commordore Amiga infamy, stating something like, "It doesn't need a faster more powerful CPU, they only play games on it". Then they all stood around in shock and amazement as the Amiga 600 died at the starting gate.

Or the all time classic was with Ed Esber of Asthon-Tate (the DBASE company) infamy as he single handedly destroyed the company. Chase away all the developers and you have no one left to push your product to the users. Next you put out a product that literally doesn't work becauseof all the bugs. Then to top it off Borland paid something like 400 million or so to buy the dead Ashton-Tate company.

I wonder what happened to Lotus 1-2-3? Anyone remember Visicalc anymore?

There are *many* RTOSs out there.

The one in the RCX was fair-to-middlin'. IMO.

That would be the thing to do. And it is more of a design issue than a programming issue.

I think that everybody agrees with this.

Perhaps the RoboBrix project will replace this.

-- D. Jay Newman

It would be a shame to lose the LEGO Mindstorms. Everyone complains that it is too expensive. Compared to what? Where can you get a robot kit that requires no soldering, has a graphical (or not if you use NQC) programming IDE, has hundreds, if not thousands of optional ways to build it, and just about everyone can understand how it works, finally, that costs less than $200? I've seen plenty of other student robot kits and products, not a single one comes in at less than $500. Mindstorms was as cheap as you can get. Sure I can design a robot for $15. But can a 10 year old build it? I taught a class on robotics to 16 kids from 11 to 16 years old using AVR 2313 board that I designed, Mark III chassis, and Bascom AVR. It was tough, but they all got through it. I could have done more with Mindstorms, there is a limit on what you can start with. Also, plenty of folks designed even more sensors for the Mindstorms that gave it compasses, IR proximity detection, heat, pressure and sound detection... Pretty damned impressive. It is easy to knock the expense of the LEGO product, but quality and innovation does NOT come for free. Should they give their product away for the joy of working on the designs? The only thing I see from MegaBlocks is cheap parts and lackluster designs. I usually gaze in wonder at what the LEGO engineers come up with. The "get it cheap" mentality is driving our (USA) economy into the toilet as folks strive to buy cheaper and cheaper junk, eventually _someone_ is going to demand quality again, aren't they?

Done with soapbox, IMO regardless, DLC

I can't speak for the others, but to me, it's not a question of what you get for the money, which is indeed spectacular, but that the RIS package costs $200. They could double the content and charge $300, and it would still be "too expensive." It's not what you get that always matters, but simply how much it costs.

Could they, for example, have put out a $129 "starter package" that included just the RCX, tower, and two motors? They might then have offered truly useful enhancement packages (not the silly Robosports or Mars explorer packages, at $50 each) with more parts, switch sensors, light sensors, and so on. And not just from Shop-at-Home, but local retailers.

It *is* hard to match a Mercedes SL500 considering everything you get. But not everyone wants to pay $90K for a car. So Mercedes offers cheaper models with a different selection of features. Couldn't LEGO have done something like this with the Mindstorms?

What they *did* do was come out with products that didn't leverage off one another. True, the Mini-scout could be programmed from an RCX using flashing light, but LEGO made this information hard to come by. The Scout could be programmed using the RCX tower, but except for a beta SDK LEGO once offered, there was hardly any note of this.

-- Gordon Author: Constructing Robot Bases, Robot Builder's Sourcebook, Robot Builder's Bonanza

Hi Gorden,

I think what we have here is a Half-Full person (Dennis) and a Half-Empty person. (You) You seem to be concentrating on the negative aspects of Lego Mindstorms while Dennis is seeing its positive side.

The Lego Company may have their corporate head up their ass, but I still find Lego Mindstorms system FANTASTIC. Fantastic for what I see it being designed for, children. Adults can use it, just as adults, me included, can have fun playing with their Original Lego products but it's for kids, for beginners.

As far as cost , if it's so overpriced then why isn't there any competition? Apple Computers screwed up in the marketing of their product and IBM took them to the cleaners. IBM screwed up with their marketing of the PC and the clones took over. Why hasn't somebody made cheaper / better version of the Mindstorms?

Let me ask again, what else is there? Where is the competition?

Jay

---------------------------------------------------------------------------- "I'm pullin' for you; we're all in this together", Red Green

----------------------------------------------------------------------------

Define "competition." Exactly like Mindstorms=0. Some other robot platform=lots. Seeing how Mindstorms sales have dropped off (they peaked in 2001), this either means Mindstorms reached saturation, or other offerings eventually took over the space Mindstorms once occupied. Or both.

I'm not being negative about Mindstorms, but an observer looking at what might have been. I pointed out that $200 is a lot of money for many people, regardless of what you get for it. You read other messages in these threads were people complain Mindstorms was expensive. It was; they didn't say it wasn't a good *value*. Two different concepts here.

You might wonder if LEGO could have provided a lower cost entry-level product that could have been scaled up to the full RIS kit. They didn't do it this way. Instead they offered all sorts of cheaper products that were only marginally upgradable. Even after several iterations of the RCX, even after seeing the phenomenal interest by more advanced users, they didn't provide a simple port on the thing for expanding it. That would have added maybe $2. Yet they might have even been able to charge $20 more for the RIS, and provided an upgrade path to keep users going.

*IF* Mindstorms goes away, and assuming the toy market comes back to life, we'll probably see a directly similar competitor to Mindstorms. From whom I don't know. Until then, whoever comes out with the product can study the market and determine how to do it better.

-- Gordon Author: Constructing Robot Bases, Robot Builder's Sourcebook, Robot Builder's Bonanza

I feel that that they never provided a way for people to expand it. You had three motor outputs and three sensor I/O's and that was it. The kid as a rank beginner, quickly outgrows it, and loses interest. What else can they do with it? After you build a few robots (that you can with what you got), what is there left to do? Thermometer? ah heck I don't have a temperature sensor. And I can't buy one either. Oooohhh I can build a third arm like thing, ah heck I need a third motor, bummer they want $79.00 for the accessory kit with a motor in it. My dad won't buy it for me right now, later you find out that accessory kit looks like it was discontinued no one has it in stock anymore. (Of course that could have been the classic Nintendo like scheme to drive up the prices with exclusivity and demand). Then a lot of the manual examples depict robots that require other accessory kits to complete, except you can't get the kits.. The kids lose interest quickly. There is no path for the users as they advance to advance to. The GAP from the LEGO Mindstorms to the other robot kits is pretty big. You mainly have to learn electronic, soldering, and more advanced programming skills. Which isn't good for the kids.

Legos only needed a way to bring out 16 I/O pins so we could access and program the RCX to use them. Then adding a few more sensors like Ultrasound, et cetera would have helped a lot too. As it is, you only have the bumper swtich, light sensor, and the rare temperature sensor. Things like the Sharp IR object sensor's would have been really good. Or maybe a break out Lego block to allow hooking up most any sensor to it.

They already had the programming languages in the form of the graphical system, Mindscripts and NQC. They just never left a path for users to advance into. Kids lose interest fast. "Older kids" get frustrated.

Kids needed examples and role models to motivate them. When they see a contest or a group doing neat things with the Mindstorms, they keep their interest and want to do more. The "First" groups were a good start, but they seem to be restricted thus locking out others just getting in. They really needed more Legos Robotics clubs and such. But I just don't see it happening.

Earl,

I totally agree with you here. LEGO never provided for expansion, the RCX was clearly capable, heck all of the sensors you mentioned have been made by someone for the RCX. A simple edge card expansion on the brick with an optional expander would have been simple and fab. But, LEGO seemed to release the Mindstorms, then ignored it. After all that effort they so totally dropped the ball that it landed on their feet, broke all their toes and sent them to the hospital...

I don't think that it was overpriced, I think that it was undersupported!

IMO, DLC

: I feel that that they never provided a way for people to expand it. : You had three motor outputs and three sensor I/O's and that was it. : The kid as a rank beginner, quickly outgrows it, and loses interest. What : else can they do with it? : After you build a few robots (that you can with what you got), what is : there left to do? Thermometer? ah heck I don't have a temperature sensor. : And I can't buy one either. Oooohhh I can build a third arm like thing, ah : heck I need a third motor, bummer they want $79.00 for the accessory kit : with a motor in it. My dad won't buy it for me right now, later you find out : that accessory kit looks like it was discontinued no one has it in stock : anymore. (Of course that could have been the classic Nintendo like scheme to : drive up the prices with exclusivity and demand). : Then a lot of the manual examples depict robots that require other : accessory kits to complete, except you can't get the kits.. : The kids lose interest quickly. : There is no path for the users as they advance to advance to. : The GAP from the LEGO Mindstorms to the other robot kits is pretty big. You : mainly have to learn electronic, soldering, : and more advanced programming skills. Which isn't good for the kids.

: Legos only needed a way to bring out 16 I/O pins so we could access and : program the RCX to use them. : Then adding a few more sensors like Ultrasound, et cetera would have helped : a lot too. As it is, you only have the bumper swtich, light sensor, and the : rare temperature sensor. Things like the Sharp IR object sensor's would have : been really good. : Or maybe a break out Lego block to allow hooking up most any sensor to it.

: They already had the programming languages in the form of the graphical : system, Mindscripts and NQC. : They just never left a path for users to advance into. : Kids lose interest fast. "Older kids" get frustrated.

: Kids needed examples and role models to motivate them. When they see a : contest or a group doing neat things with the Mindstorms, they keep their : interest and want to do more. : The "First" groups were a good start, but they seem to be restricted thus : locking out others just getting in. They really needed more Legos Robotics : clubs and such. But I just don't see it happening.

:> Define "competition." Exactly like Mindstorms=0. Some other robot :> platform=lots. Seeing how Mindstorms sales have dropped off (they peaked :> in 2001), this either means Mindstorms reached saturation, or other :> offerings eventually took over the space Mindstorms once occupied. Or :> both. :>

:> I'm not being negative about Mindstorms, but an observer looking at what :> might have been. I pointed out that $200 is a lot of money for many :> people, regardless of what you get for it. You read other messages in :> these threads were people complain Mindstorms was expensive. It was; :> they didn't say it wasn't a good *value*. Two different concepts here. :>

:> You might wonder if LEGO could have provided a lower cost entry-level :> product that could have been scaled up to the full RIS kit. They didn't :> do it this way. Instead they offered all sorts of cheaper products that :> were only marginally upgradable. Even after several iterations of the :> RCX, even after seeing the phenomenal interest by more advanced users, :> they didn't provide a simple port on the thing for expanding it. That :> would have added maybe $2. Yet they might have even been able to charge :> $20 more for the RIS, and provided an upgrade path to keep users going. :>

:> *IF* Mindstorms goes away, and assuming the toy market comes back to :> life, we'll probably see a directly similar competitor to Mindstorms. :> From whom I don't know. Until then, whoever comes out with the product :> can study the market and determine how to do it better. :>

:> -- Gordon :> Author: Constructing Robot Bases, :> Robot Builder's Sourcebook, Robot Builder's Bonanza

I built this thing in a way that can be replicated almost without soldering:

motor controller was built from a kit but can be bought complete.) The only soldering required is connecting the tags on the motor to some wire but once you've done that they can be reused for other projects. The body is made from scrap acrylic sheet (bought for $1/lb) and the hardest part of the construction was drilling holes for screws, something I think a kid could to without power tools. Cutting the acrylic is just a matter of scoring and snapping it. The rest of the connections are just breadboard type connections and a bit of glue. And the microcontroller board cost about $30. Programming it was non-trivial but the construction was something any reasonably adpet kid could have done. To be honest I built it faster than any Mindstorms robot I have built.

-- Torque

Torquemada wrote: In other words, nothing even close to Mindstorms. ;)

DLC

:> When I say reasonable, what I mean is some robotic system :> which has a similar flexibility and learning curve as Mindstorms and :> *no* soldering whatsoever...

: I built this thing in a way that can be replicated almost without soldering: :

(The motor controller was built from a kit but can be bought complete.) : The only soldering required is connecting the tags on the motor to some : wire but once you've done that they can be reused for other projects. : The body is made from scrap acrylic sheet (bought for $1/lb) and the : hardest part of the construction was drilling holes for screws, something : I think a kid could to without power tools. Cutting the acrylic is just a : matter of scoring and snapping it. The rest of the connections are just : breadboard type connections and a bit of glue. And the microcontroller : board cost about $30. Programming it was non-trivial but the construction : was something any reasonably adpet kid could have done. To be honest I : built it faster than any Mindstorms robot I have built. : -- : Torque